How I use agent loops and goals (Claude Code + Codex).

🔗 Resources and example prompts: https://github.com/owainlewis/youtube-tutorials/tree/main/tutorials/agent-goals-and-loops 💬 Community: https://aiengineer.co In this video I walk through how I use agent loops, goals, and scheduled automations in Claude Code and Codex for real AI coding work. I break down the difference between prompts, goals, loops, and schedules, then show practical examples from my own workflow: scanning production logs, deploying an app to Google Cloud, working through GitHub issues, triaging a backlog, and keeping documentation in sync with code. This is not about vague “agentic” hype. It is a practical guide to using loop engineering and goal-oriented workflows to get more useful, reliable work out of AI coding agents. What you'll learn: - What agent loops and goals actually mean in Claude Code and Codex - How Codex and Claude Code approach goals differently - How to use scheduled automations for maintenance and monitoring - When goals are useful for long-running coding tasks - How to structure success criteria so an agent can verify its own work - Where these workflows make sense, and where they are too risky - How to think about cost, token usage, and the hype around loop engineering 🔗 Links Resources and example prompts: https://github.com/owainlewis/youtube-tutorials/tree/main/tutorials/agent-goals-and-loops Community: https://aiengineer.co If this was useful, a like and subscribe genuinely helps the channel. Drop a comment with any Claude Code or Codex workflows you want me to cover next. #ClaudeCode #Codex #OpenAICodex #AICoding #AIAgents #AIEngineering #SoftwareEngineering #AgenticCoding